JP2009206800A - Image processing apparatus, projector and image processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve distortion correction accuracy of a projected image, the distortion occurring in projection of an image to a screen on which wrinkles and slack have arisen. <P>SOLUTION: If an optical axis of a photographing part 20 is inclined even when a projection correction pattern image 220 is rectangularly displayed on a projection screen SC, trapezoidal distortion occurs in a photography correction pattern image 230. From this cause, a projector 10 generates a transform expression Φ of projection transform which is used for trapezoidal distortion correction processing by using a correction pattern image 210 and the photography correction pattern image 230. By means of the transform expression Φ of projection transform, a coefficient is calculated by applying a least square approximation method so that the square sum of differences between the coordinates Φ[Qi(xi, yi)] of each feature point obtained by converting, by means of the transform expression Φ, the coordinates Qi(xi, yi) of each feature point of the projection correction pattern image 220 on which curved surface distortion has occurred and the coordinates Q1(xi, yi) of each feature point of the correction pattern image 210 having no distortion is minimized. The curved surface distortion is corrected after correcting the trapezoidal distortion by using the transform expression Φ, thereby correction accuracy can be improved. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to image processing for a projector, and more particularly to a technique for correcting distortion of an image displayed on a projection surface.

  An image display system that displays an image on a projection surface using an image display device such as a projector is used. In the image display system, for example, a temporary projection surface that is installed only when the image display system is used and stored when not used is used. Such a temporary projection surface is likely to be loose, and a displayed image may be distorted. Therefore, a pattern image in which dot figures are arranged at the same interval is projected on the screen, and the projected image is shot with a shooting device such as a digital camera installed at the viewer's viewpoint position. Disclosed is a technique for performing distortion correction on an image based on a deviation from a position where a point graphic is shown in the original pattern image, and projecting an image without distortion on a screen.

JP 2001-083949 A

  However, when a pattern image displayed on the screen is photographed with the optical axis of the photographing apparatus tilted with respect to the screen normal, the trapezoidal distortion corresponding to the inclination of the photographing apparatus occurs in the photographed pattern image. . When curved surface distortion correction is performed using a pattern image in which trapezoidal distortion occurs in this way, trapezoidal distortion in the direction opposite to that generated in the captured pattern image occurs in the projected image after correction. There is a problem that the accuracy of curved surface distortion correction is lowered.

  The present invention has been made in view of the above-described problems, and an object of the present invention is to improve the accuracy of distortion correction of a projected image when an image is projected onto a screen where wrinkles or slack are generated.

  SUMMARY An advantage of some aspects of the invention is to solve at least a part of the problems described above, and the invention can be implemented as the following forms or application examples.

[Application Example 1]
An image processing apparatus for correcting distortion of a projected image projected on a projection plane by a projector, the photographing correction image representing the correction image projected on the projection plane by the projector and photographed by a photographing apparatus Based on the shooting correction image data acquisition means for acquiring data, the correction image data acquisition means for acquiring correction image data representing the correction image, the shooting correction image data, and the correction image data. Trapezoidal distortion correction means for performing trapezoidal distortion correction processing for correcting the trapezoidal distortion generated in the imaging correction image represented by the imaging correction image data due to the inclination of the imaging apparatus to the imaging correction image data; The trapezoidal corrected image in which the keystone distortion of the shooting correction image is corrected and the correction image are compared, and the distortion is caused by the distortion of the projection plane. Correction information generating means for generating curved surface distortion correction information for correcting curved surface distortion generated in a copy image, projection image data acquiring means for acquiring image data representing an image projected on the projection surface, and curved surface distortion correction An image processing apparatus comprising: a curved surface distortion correction unit that performs a curved surface distortion correction process on the acquired image data using information.

  According to the image processing apparatus of Application Example 1, since the trapezoidal distortion of the shooting correction image is corrected before the curved surface distortion correction processing, the curved surface distortion correction processing is performed using the imaging correction image in which the trapezoidal distortion does not occur. The curved surface distortion correction information to be used can be generated. Therefore, even if trapezoidal distortion occurs in the image for correction of photographing due to the inclination of the optical axis of the photographing means, curved surface distortion generated in the projected image due to distortion of the projection surface can be accurately corrected. Therefore, the image quality of the image projected on the projection surface can be improved.

In the image processing apparatus according to the application example 1, the coefficient for minimizing the difference between both sides of the conversion formula of Formula 1 is calculated using the photographing correction image data and the correction image data, and the calculated coefficient is applied. The trapezoidal distortion correction process is performed on the photographing correction image data using the conversion formula.

  According to the image processing apparatus of the application example 1, the constant is determined so that the difference between both sides of the conversion formula used for the trapezoidal distortion correction process is minimized. Therefore, it is possible to accurately correct the trapezoidal distortion of the image data for photographing correction including the curved surface distortion.

  In the image processing apparatus according to the application example 1, the correction image represented by the correction image data includes a plurality of feature points represented by a predetermined figure, and the trapezoidal distortion correction unit includes the photographing correction. Least square approximation so that the sum of squares of the difference between the position obtained by converting the position of the plurality of feature points included in the image for conversion by the conversion formula and the position of the plurality of feature points of the correction image is minimized. The coefficient of the conversion formula is calculated using the method. According to the image processing apparatus of the application example 1, the constant can be calculated using the least square method. Therefore, the accuracy of the constant can be improved, and the accuracy of the trapezoidal distortion correction of the captured image data including the curved surface distortion can be improved.

In the image processing apparatus of Application Example 1, the trapezoidal distortion correction unit calculates the coefficient of the conversion equation by applying the equation of Equation 2.

  According to the image processing apparatus of the application example 1, it is possible to correct the trapezoidal distortion of the photographing correction pattern image using at least four feature points included in the correction image. Therefore, the constant can be calculated with arbitrary accuracy by increasing or decreasing the number of feature points to be used in accordance with the performance of the image processing apparatus and the accuracy of correction desired by the user.

[Application Example 2]
A projector, which is a light source, an image forming unit that modulates light emitted from the light source based on image data, and forms an image, the image processing apparatus of application example 1, and the corrected image data that is the image forming unit A correction image data input means for inputting to the projector, and a corrected image data input means for inputting the image data subjected to the curved surface distortion correction processing by the image processing apparatus to the image forming section. According to the projector of Application Example 2, curved surface distortion correction information is generated using a captured image in which trapezoidal distortion is corrected, and based on image data that has been subjected to curved surface distortion correction processing using the generated curved surface distortion correction information. Can project images. Therefore, even if trapezoidal distortion occurs in the photographing correction image due to the inclination of the optical axis of the photographing means, an image without distortion can be displayed on the projection plane.

  In the present invention, the various aspects described above can be applied by appropriately combining or omitting some of them. In addition to the configuration as the image processing apparatus described above, the present invention records an image processing method by the image processing apparatus, a computer program for causing the image processing apparatus to execute image processing, and the computer program recorded in a computer-readable manner. It can also be configured as a recording medium. In any configuration, the above-described aspects can be appropriately applied. As a computer-readable recording medium, various media such as a flexible disk, a CD-ROM, a DVD-ROM, a magneto-optical disk, an IC card, and a hard disk can be used.

A. Example:
A1. System configuration:
FIG. 1 is an explanatory diagram illustrating an image display system in the embodiment. FIG. 1A shows the system configuration of the image display system of the embodiment, and FIGS. 1B and 1C show the relationship between the optical axis of the photographing unit 20 and the screen normal. Yes. The image display system includes a projector 10, a photographing unit 20, and a projection surface SC. The projection screen SC is a hanging screen.

  The photographing unit 20 is a digital camera, photographs a projection image projected on the projection surface SC, and inputs photographed image data to the projector 10. In the embodiment, the photographing unit 20 photographs a projection correction pattern image 220 that is a correction pattern image projected on the projection plane by the projector 10. The correction pattern image is an image in which a plurality of feature points P represented by point graphics are regularly arranged at a predetermined interval, and is partially included in the image due to wrinkles or loosening of the projection surface SC. It is used to correct distortion that occurs in In addition, arrangement | positioning of the feature point P is an example, and is not restricted to this. Hereinafter, in the embodiment, distortion generated in an image due to a bulge or depression caused by wrinkles or slack of the projection surface SC is referred to as curved surface distortion, and a correction pattern image captured by the imaging unit 20 is referred to as a shooting correction pattern image 230. In the correction pattern image 220 displayed on the projection screen SC, the keystone distortion caused by the tilt of the projector 10 is corrected in advance by the user (or projector).

  By the way, when the projection correction pattern image 220 is photographed, it is preferably photographed from the viewpoint position of the viewer, and the optical axis La of the photographing unit 20 and the normal line of the projection plane SC are preferably substantially parallel. However, since the user captures the projection correction pattern image 220 with the photographing unit 20 in his / her hand, the photographing correction pattern image 230 has a table corresponding to the inclination of the photographing unit 20 as shown in FIG. Shape distortion occurs. Specifically, as shown in FIG. 1B, the photographing unit 20 photographs the projection plane SC in a state where the photographing unit 20 is rotated about the lens optical axis La of the photographing unit 20, as shown in FIG. As shown in FIG. 6, when the projection surface SC is photographed in a state where the optical axis LA of the photographing unit 20 is inclined with respect to the screen normal H, trapezoidal distortion occurs in the photographing correction pattern image. When the curved surface distortion correction is performed using the photographing correction pattern image in which the trapezoidal distortion is generated, the accuracy of the curved surface distortion correction is lowered. Hereinafter, in the embodiment, unless otherwise specified, the projection surface SC is photographed while being rotated with the lens optical axis La of the photographing unit 20 as the rotation axis, or the optical axis LA of the imaging unit is set to the screen normal H. A trapezoidal distortion that occurs in a photographic correction pattern image by shooting the projection surface SC in an inclined state is called a trapezoidal distortion.

  Therefore, in the embodiment, the projector 10 corrects the trapezoidal distortion of the shooting correction pattern image 230 before the curved surface distortion correction process, and the shooting correction pattern image in which the trapezoidal distortion is corrected (hereinafter referred to as a trapezoid correction image in the embodiment). ) Is used to correct the curved surface distortion generated in the projected image due to the curved surface distortion of the projection surface SC.

A2. Function block:
FIG. 2 is an explanatory diagram illustrating functional blocks of the projector 10 in the embodiment. The projector 10 includes a light source lamp 11, a liquid crystal light valve 13, and a projection optical system 15. In FIG. 2, each optical system is shown in a simplified manner. The projector 10 includes a light source lamp driving unit 12, a liquid crystal light valve driving unit 14, a projection optical system adjusting unit 16, an image processing unit 50, and an operation unit 130. The image processing unit 50 includes a CPU 100, an image acquisition unit 102, a trapezoidal distortion correction information generation unit 104, a trapezoidal distortion correction unit 106, a curved surface distortion correction information generation unit 108, a curved surface distortion correction unit 110, and a memory 120. The CPU 100 controls the overall operation of the projector 10. In the embodiment, the image processing unit 50 corresponds to an “image processing apparatus” in the claims.

  The light source lamp driving unit 12 drives the light source lamp 11. The liquid crystal light valve driving unit 14 drives the liquid crystal light valve 13 according to the image data given from the image processing unit 50.

  The operation unit 130 causes the CPU 100 to execute various processes in response to input from the user. For example, the user can create correction information used for the distortion correction processing or adjust the contrast of the image projected on the projection screen SC via the operation unit 130.

  The projection optical system adjustment unit 16 adjusts the position of the projection optical system 15. Specifically, the projection optical system adjustment unit 16 moves the lens included in the projection optical system 15 or adjusts the shift position in a direction parallel to the light source optical axis LA. The light source optical axis represents the central axis of light emitted from the light source lamp 11. Note that the projection optical system adjustment unit may not be provided, and the position of the projection optical system may be manually adjusted.

  In the memory 120, correction pattern image data 200 used for distortion correction of the projected image is stored in advance. The correction pattern image data 200 is generated in advance so as to efficiently detect the distortion of the projected image according to the design of the optical system of the projector.

  The image acquisition unit 102 has a function of transferring the correction pattern image data 200 or image data given from the outside to the image processing unit 50, and is configured by an A / D converter, for example. In the embodiment, the image acquisition unit 102 acquires image data to be projected on the projection surface from the memory card MC. The image acquisition unit 102 includes functions realized by “correction image data input means” and “correction image data input means” in the claims.

  The trapezoidal distortion correction information generation unit 104 corrects the trapezoidal distortion of the shooting correction pattern image 230 using the shooting correction pattern image data representing the shooting correction pattern image 230 and the correction pattern image data 200 stored in the memory 120. A conversion formula for generating

  The trapezoidal distortion correction unit 106 performs trapezoidal distortion correction processing on the shooting correction pattern image data using the conversion formula generated by the trapezoidal distortion correction information generation unit 104, and the trapezoidal distortion of the shooting correction pattern image is corrected. Keystone correction image data representing a keystone correction image is generated and transferred to the curved surface distortion correction information generation unit 108. The trapezoidal distortion correction unit 106 is configured by, for example, an image processor.

  The curved surface distortion correction information generation unit 108 uses the trapezoidal correction image data and the correction pattern image data 200 stored in the memory 120 to correct the curved surface distortion correction used for correcting the distortion of the projected image due to the curved surface distortion of the projection surface SC. Generate information.

  The curved surface distortion correction unit 110 processes the image data delivered from the image acquisition unit 102 using the curved surface distortion correction information calculated by the curved surface distortion correction information generation unit 108, generates corrected image data, and generates a liquid crystal light. Delivered to the valve drive unit 14. The curved surface distortion correction unit 110 is configured by, for example, an image processor.

  The liquid crystal light valve drive unit 14 forms an image for distortion correction on the liquid crystal light valve 13 in accordance with the corrected image data for driving the liquid crystal light valve 13 using the corrected image data transferred from the curved surface distortion correction unit 110. Then, an image without curved surface distortion is displayed on the screen SC. The image for distortion correction is an image distorted as a result in order to cancel out distortion caused by bulging due to wrinkles or slack of the projection surface SC. In this embodiment, the liquid crystal light valve 13, the liquid crystal light valve driving unit 14, the projection optical system 15, and the projection optical system adjustment unit 16 correspond to the “image forming unit” in the claims.

  Note that part of the functions realized by the projector 10 (for example, the functions of the image processing unit 50) may be realized by a personal computer.

A3. Distortion correction processing:
The distortion correction process will be described with reference to FIGS. FIG. 3 is a flowchart for explaining the distortion correction processing in the embodiment. FIG. 4 is an explanatory diagram for explaining trapezoidal distortion correction processing in the embodiment. FIG. 5 is an explanatory diagram illustrating curved surface distortion correction information generation processing in the embodiment. FIG. 6 is an explanatory diagram for explaining curved surface distortion correction processing in the embodiment. The distortion correction process is started when a user inputs a distortion correction process start instruction via the operation unit 130.

  The CPU 100 acquires the correction pattern image data 200 from the memory 120 and projects it on the projection surface SC (step S100). In response to a user operation, the imaging unit 20 captures a correction pattern image projected on the projection surface SC and generates captured correction image data.

  The CPU 100 acquires the shooting correction pattern image generated by the shooting unit 20 (step S102), and generates a conversion formula Φ for correcting the trapezoidal distortion of the shooting correction pattern image using the least square approximation method (step S104). ).

  A trapezoidal distortion correction process will be described with reference to FIG. In FIG. 4, a correction pattern image 210 indicates an image displayed on the liquid crystal light valve 13 in accordance with the correction pattern image data 200. The projection correction pattern image 220 represents an image obtained by projecting the correction pattern image 210 displayed on the liquid crystal light valve 13 onto the projection surface SC. The shooting correction pattern image 230 represents the projection correction pattern image 220 shot by the shooting unit 20. In the embodiment, the correction pattern image 210 displayed on the liquid crystal light valve 13 is represented by a two-dimensional XY coordinate space, and the positions of the pixels constituting the correction pattern image 210 are represented by XY coordinates. The position of each feature point P is represented by the coordinates (X, Y) of the pixel where each feature point P is represented. Henceforth, when expressing a predetermined feature point, it represents with Pi (Xi, Yi) using the variable i. The variable i is an integer of 1 or more. Further, the positions of the pixels constituting the photographing correction pattern image 230 are represented by xy coordinates. The position of each feature point Q is represented by the coordinates (x, y) of the pixel where each feature point Q is represented. The feature point Qi (xi, yi) of the shooting correction pattern image 230 corresponds to the feature point Pi (Xi, Yi) of the correction pattern image 210.

  As shown in the shooting correction pattern image 230 in FIG. 4, even if the projection correction pattern image 220 is displayed in a rectangular shape on the projection surface SC, if the optical axis of the shooting unit 20 is tilted, the shooting correction pattern image 230 is displayed in the shooting correction pattern image 230. , Trapezoidal distortion occurs. The CPU 100 uses the correction pattern image 210 and the photographing correction pattern image 230 to generate a projection transformation conversion formula Φ used in the trapezoidal distortion correction process, as shown in Equation 1. The transformation formula Φ of the projective transformation is obtained by converting the coordinates Q1 (x1, y1), Q2 (x2, y2)... Qn (xn, yn) of each feature point of the projection correction pattern image 220 in which the curved surface distortion is caused by the transformation formula Φ. Φ [Q1 (x1, y1)], Φ [Q2 (x2, y2)]... Φ [Qn (xn, yn)], which are the coordinates of each feature point after conversion, and a correction pattern image 210 without distortion. Are applied by the least square approximation method so that the sum of squares of the differences from the coordinates P1 (X1, Y1), P2 (X2, Y2). It is the formula which calculated B, C, D, E, F, G, and H.

  Specifically, the CPU 100 detects the coordinates (Xi, Yi) of each feature point Pi of the correction pattern image 210 and the coordinates (xi, yi) of each feature point Qi of the shooting correction pattern image 230, and the equation 2 8 are created, and the coefficients A, B, C, D, E, F, G, and H of the transformation formula Φ of the projective transformation are calculated. The 8-dimensional simultaneous equations of Equation 2 can be solved by applying an 8 × 8 inverse matrix on the left side from the left. In order to calculate the coefficients A, B, C, D, E, F, G, and H, at least four feature points Pi and four feature points Qi corresponding to these four feature points Pi are required. . Since one point has two pieces of information, the x component and the y component, if there are four points, the conditions of 4 points × 2 components = 8 components are met, and the conversion formula can be calculated. According to the characteristics of the least square method, the accuracy of the coefficients A, B, C, D, E, F, G, and H improves as the number of feature points used in the calculation increases.

  The CPU 100 performs trapezoidal distortion correction processing on the photographing correction pattern image 230 using the conversion formula Φ generated in this manner, and generates keystone corrected image data (step S106). Specifically, each pixel of the photographing correction pattern image 230 is converted by the conversion formula Φ. A trapezoidal correction image 240 shown in FIG. 4 is an image obtained by converting each pixel of the photographing correction pattern image 230 with the conversion formula Φ. The trapezoidal correction image 240 is represented by the same XY coordinate space as the correction pattern image 210. As shown in FIG. 4, in the trapezoidal correction image 240, the trapezoidal distortion represented in the shooting correction pattern image 230 is corrected. Each feature point R of the trapezoid correction image 240 is represented by an XY coordinate system obtained by converting each feature point Q of the photographing correction pattern image 230 by the conversion formula Φ. In this embodiment, the coordinates of the feature point R in the XY coordinate system are expressed as Ri (ai, bi) = Φ [Qi (xi, yi)]. Thus, the trapezoidal distortion correction process for the shooting correction pattern image 230 is completed.

  When the trapezoidal distortion correction process ends, the CPU 100 performs a curved surface distortion correction process. The CPU 100 uses the trapezoid correction image data representing the trapezoid correction image 240 in which the trapezoid distortion of the shooting correction pattern image 230 is corrected, to generate curved surface distortion correction information that gives a distortion in the opposite direction to the distortion generated in the projected image (step). S108).

  Generation of distortion correction information will be described with reference to FIGS. FIG. 5A shows the correction pattern image 210 displayed on the light valve, and FIG. 5B shows the trapezoid correction image 240. In FIG. 5A, feature points P1, P2,... Pn represent feature points of the correction pattern image 210. In FIG. 5B, feature points R1, R2,... Rn indicate feature points of the trapezoid correction image 240. The feature points R1, R2,... Rn correspond to the feature points P1, P2,. Since the correction pattern image 210 and the trapezoid correction image 240 are expressed in the same XY coordinate system, the difference between the coordinate components of the feature points P1, P2,... Pn and the feature points R1, R2,. Amount.

  As shown in FIG. 5B, in the trapezoidal correction image 240, the feature points R1, R2,... Rn are shifted from the feature points P1, P2,. For example, the feature point R2 is displayed as being shifted upward from the feature point P2. In order to display each feature point Ri at the ideal projection position pi, as shown in FIG. 6A, the position of each feature point Pi of the correction pattern image 210 corresponding to each feature point Ri is set to the calculated distortion amount. Accordingly, the feature points Ri may be moved in the direction opposite to the distortion direction.

  The CPU 100 analyzes the trapezoid correction image 240 to detect the coordinates (ai, bi) of the feature points Ri, and the positions of all the feature points Ri correspond to the feature points Ri (Xi, Yi). The amount of distortion indicating how much is deviated from which direction is calculated. The amount of distortion may be, for example, the difference (ai−Xi, bi−Yi) between the coordinates of the feature point Pi and the coordinates of the feature point Ri. In this case, for example, the distortion amount of the feature point R2 can be expressed as a difference (a2-X2, b2-Y2) between the coordinates of the feature point P2 and the coordinates of the feature point R2. The CPU 100 generates distortion correction information in which the distortion amounts of all feature points calculated in this way are collected in a table.

  The CPU 100 uses the generated distortion correction information to perform a distortion correction process that gives distortion in the opposite direction to the distortion of the feature point on the input image data (step S16), and based on the corrected image data that has been subjected to the distortion correction process. Then, an image for distortion correction is displayed on the light valve 13. (Step S18). In addition, when performing distortion correction processing, not only is distortion applied only to the location corresponding to each feature point, but also the area around each feature point is, for example, the amount of distortion between adjacent feature points. It is preferable to complement the distortion by approximating based on it. FIG. 6B shows, as an example, a distortion correction image 250 displayed on the light valve 13 based on the corrected image data subjected to the distortion correction processing. As shown in FIG. 6B, each feature point Pi of the distortion correction image 250 is moved in a direction opposite to the distortion of the feature point Ri of the trapezoid correction image 240 according to the distortion correction information. As a result, an image without distortion is displayed on the projection surface SC.

  According to the projector 10 of the embodiment, the trapezoidal distortion correction process can be performed on the shooting correction pattern image 230 before the curved surface distortion correction process is performed using the shooting correction pattern image 230. Then, using the trapezoidal correction image 240 in which the trapezoidal distortion of the photographing correction pattern image 230 is corrected, curved surface distortion correction information for correcting the curved surface distortion generated in the projection image due to the curved surface distortion of the projection surface SC is generated. Accordingly, even when a trapezoidal distortion occurs in the photographing correction pattern image 230 due to the inclination of the optical axis of the photographing unit 20 when photographing the projection correction pattern image, it is possible to accurately detect the curved surface distortion generated in the projected image. Therefore, the accuracy of curved surface distortion correction can be improved, and an image without distortion can be projected on the projection surface.

  Further, according to the projector 10 of the embodiment, since the conversion formula Φ is calculated using the least square method, the trapezoidal distortion of the photographing correction pattern image 230 can be accurately corrected. Therefore, even when the optical axis of the photographing unit 20 is inclined, highly accurate curved surface distortion correction information can be generated from the photographing correction pattern image 230. Therefore, equipment such as a tripod for fixing the photographing unit 20 and equipment for adjusting the optical axis of the photographing unit 20 are not required, and the portability of the image display system can be increased. In addition, the cost can be reduced by making these equipment preparations unnecessary.

  Further, according to the projector 10 of the embodiment, since the adjustment process of the optical axis of the photographing unit 20 is not necessary, the setup time can be shortened.

B. Variations:
(1) In the embodiment, feature points are regularly arranged in the correction pattern image 210, but the feature points may be randomly arranged. In addition, the feature points are not limited to circular figures, and characters and other figures may be used.

(2) In the embodiment, the correction pattern image data 200 is stored in advance. However, for example, the correction pattern image data 200 may be acquired from another device, or according to the design value of the optical system of the projector 10. The correction pattern image data 200 may be created as appropriate. This saves memory capacity.

  As mentioned above, although the various Example of this invention was described, this invention is not limited to these Examples, A various structure can be taken in the range which does not deviate from the meaning.

It is explanatory drawing which illustrates the image display system in an Example. It is explanatory drawing which illustrates the functional block of the projector 10 in an Example. It is a flowchart explaining the distortion correction process in an Example. It is explanatory drawing explaining the trapezoid distortion correction process in an Example. It is explanatory drawing explaining the curved surface distortion correction information generation process in an Example. It is explanatory drawing explaining the curved surface distortion correction process in an Example.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Projector 11 ... Light source lamp 12 ... Light source lamp drive part 13 ... Liquid crystal light valve 13 ... Light valve 14 ... Liquid crystal light valve drive part 15 ... Projection optical system 16 ... Projection optical system adjustment part 20 ... Shooting part 50 ... Image processing part 100 ... CPU
DESCRIPTION OF SYMBOLS 102 ... Image acquisition part 104 ... Trapezoid distortion correction information generation part 106 ... Trapezoid distortion correction part 108 ... Curved surface distortion correction information generation part 110 ... Curved surface distortion correction part 120 ... Memory 130 ... Operation part 200 ... Correction pattern image data 210 ... Correction pattern Image 220 ... Projection correction pattern image 230 ... Shooting correction pattern image 240 ... Trapezoid correction image

Claims (6)

An image processing apparatus that corrects distortion of a projection image projected on a projection surface by a projector using a correction image,
Shooting correction image data acquisition means for acquiring shooting correction image data representing the correction image projected on the projection plane by the projector, shot by a shooting device;
Correction image data acquisition means for acquiring correction image data representing the correction image;
Based on the shooting correction image data and the correction image data, trapezoidal distortion correction that corrects trapezoidal distortion generated in the shooting correction image represented by the shooting correction image data due to the tilt of the shooting apparatus. Trapezoidal distortion correction means for performing processing on the image data for photographing correction;
A curved surface distortion for correcting a curved surface distortion generated in the projected image due to the distortion of the projection plane by comparing the trapezoidal corrected image in which the trapezoidal distortion of the shooting correction image is corrected and the correction image. Correction information generating means for generating correction information;
Projection image data acquisition means for acquiring image data representing an image projected on the projection plane;
An image processing apparatus comprising: a curved surface distortion correction unit that performs a curved surface distortion correction process on the acquired image data using the curved surface distortion correction information. The image processing apparatus according to claim 1,
The trapezoidal distortion correction unit calculates a coefficient that minimizes the difference between both sides of the conversion formula of Formula 1 using the photographing correction image data and the correction image data, and applies the calculated coefficient. An image processing apparatus that performs the trapezoidal distortion correction processing on the photographing correction image data using a conversion formula.

The image processing apparatus according to claim 1 or 2,
The correction image represented by the correction image data includes a plurality of feature points represented by a predetermined figure,
The trapezoidal distortion correcting means is a sum of squares of differences between positions obtained by converting the positions of the plurality of feature points included in the photographing correction image by the conversion formula and positions of the plurality of feature points of the correction image. An image processing apparatus that calculates a coefficient of the conversion equation using a least-square approximation method so that is minimized. The image processing apparatus according to claim 3.
The trapezoidal distortion correction unit is an image processing apparatus that calculates the coefficient of the conversion equation by applying the equation of Equation 2.

A projector,
A light source;
An image forming unit that modulates light emitted from the light source based on image data and forms an image;
An image processing apparatus according to any one of claims 1 to 4,
Correction image data input means for inputting the corrected image data to the image forming unit;
A projector comprising: corrected image data input means for inputting image data subjected to the curved surface distortion correction processing by the image processing apparatus to the image forming unit. An image processing method executed by an image processing apparatus,
Obtaining image data for photographing correction representing the image for correction photographed on the projection plane by the projector, photographed by the photographing device;
Obtaining correction image data representing the correction image;
Based on the shooting correction image data and the correction image data, trapezoidal distortion correction that corrects trapezoidal distortion generated in the shooting correction image represented by the shooting correction image data due to the tilt of the shooting apparatus. Processing is performed on the image data for shooting correction,
A curved surface distortion for correcting a curved surface distortion generated in the projected image due to the distortion of the projection plane by comparing the trapezoidal corrected image in which the trapezoidal distortion of the shooting correction image is corrected and the correction image. Generate correction information,
Obtaining image data representing an image to be projected onto the projection surface;
An image processing method for performing curved surface distortion correction processing on the acquired image data using the curved surface distortion correction information.

Images